Determining appropriate inventory levels is one of the most important and challenging tasks that supply chain managers face. Carry too much inventory, and you have more money tied up in working capital than necessary. Carry too little, and you’ll likely run into stockouts and disrupt downstream manufacturing processes.

While those are the big-picture problems of inventory management, there are challenges in the details, too. Inventory managers must set appropriate targets for both cycle stock and safety stock — and continuously manage the levels of both. This is tricky because levels change frequently, depending on the item and industry, as orders are received, as orders are filled, and as materials are produced.

There are also challenges associated with in-process and finished product inventory. These can include:

• completed items awaiting quality test results
• extra material made to satisfy minimum lot-size or batch-size requirements
• slow-moving, obsolete and expired inventory.
  

Of course, the aggregate inventory of all of these needs to be monitored. There are some simple formulas for calculating the appropriate ranges for individual inventory items, but calculating upper and lower limits requires a bit more subjective judgment.

First, it is important to remember that the aggregate inventories should be relatively constant because some products are being produced while others are being shipped. The lower limit of aggregate inventory is calculated as the sum of safety stock; half the total cycle stock; quality inspection-held inventory; staging or transportation inventories; and held, dead or other non-performing stock, totaled for all stock keeping units (SKUs).

The allowances for the latter components can be calculated in several ways. One is simply to look at how much stock is tied up in each category. Another is to take a more analytical view by asking some questions:

• How long should it take for an item to clear quality inspection?
• How long should it take for a material to be ready for use from the time it is received?
• How long should it take us to move finished product through the plant to the shipping dock?
• How long should it take us to work off or disposition held product?
• What is the rate of products put on hold?

The answers to these questions can be converted from time to quantity by estimating an hourly, daily or weekly volume. The upper limit is more difficult to calculate and more subject to human judgment. It should be the lower limit plus some allowance for the combined variability of each SKU’s profile.

Statistically, it’s possible to calculate the combined variability of all products. However, there’s often a correlation between products and external causes, such as weather events or competitive actions that cause many products to move together, rendering the statistics invalid. Therefore, judgment, hard limits or the review of demand history may be the best methods for determining the upper limit.

Sticking to the range

The distance between the upper and lower limits should allow for inventory to bounce between them given natural variability. When the total site inventory extends beyond either of the limits, it’s unlikely that inventory will return to the range without corrective action — and that’s where good inventory management becomes critical. If too many products are low at once, it will overtax the production lines, and stockouts may result. If too many products are high at once, working capital is not being used efficiently, there may be insufficient storage space, handling and damage costs will increase, and the risks of obsolescence and shelf-life expiration increase.

The lower limit of aggregate inventory is the point at which the site can no longer produce economically because there isn’t time to make a reasonably sized batch of each product in sequence before something else becomes a priority. Disruption increases, and the shorter cycles and disruptions cause less-efficient production, which leads to even lower inventories, even shorter cycles, even less efficiency and even more disruption. 

If a site ever finds itself in this situation, the only possible ways out are to either manage demand or increase capacity. If the upper limit will be exceeded, production should be curtailed, or additional demand should be accepted — typically through promotional pricing to keep inventory in the preferred range. The latter solution requires a longer lead time and must be set in motion well before inventory limits are reached.

On the other hand, the first solution is relatively simple, but painful. It’s easy to just stop producing. But no one likes to shut down lines, especially because curtailing production often means that employees can’t meet goals or earn rewards through no fault of their own. This can be mitigated by planning in advance for what to do with downtime and using balanced measures for results and reward systems.

Aggregate inventory management in practice

The good news is that once an aggregate inventory adjustment plan is set into motion, it can be a relatively easy fix and pay dividends for years to come. Consider this example: A large integrated paper company had a wood pulp and a diaper division. One grade of internally produced wood pulp was used to make the diaper’s absorbent core. On average, the output of the pulp plant equaled the requirements of the diaper division. However, there was variation in the rate of pulp production and in the demand for diapers.

Management believed there was a cost advantage to using all the diaper pulp production internally and was reluctant to sell pulp when diaper demand was low or when the pulp division was running at high efficiency. When the pulp division had production upsets or when diaper demand was high, there was a reluctance to purchase pulp from the outside market until the last minute, resulting in crisis purchasing to avoid production shutdowns. History showed a cyclical stockpiling of large quantities of pulp in high-cost outside warehousing, followed by periods of low inventory with expedited purchases of premium-cost pulp from the outside market.

 A simulation was created to show that, if buying and selling were done mechanically at preset inventory trigger levels, the total amount of buying and selling was relatively small, inventory stayed within limits, and the overall result was more profitable. This insight convinced management that there would be times when it was better to sell excess pulp than store it and others when minimum stock should be maintained by purchasing pulp in an orderly fashion. Once these rules were followed, inventory remained within the preferred limits. Plus, the diaper lines ran better because they were supplied with fresher pulp from consistent sources.

The solution doesn’t always have to involve lots of head-scratching and major changeovers. Sometimes it just takes a few minutes of analyzing the inventory data and operations to identify where the problems lie. Even small tweaks make a difference.

Take a deeper dive into this topic with the APICS CPIM (Certified in Planning and Inventory Management) credential.